Evolution of. Plants 367 



But Henslow's able advocacy {125: 485; 126: 717) of the view 

 that monocotyledons have developed as an offshoot from the 

 protoangiospermic line, and that they have become simpli- 

 fied by the adoption of a semi-aquatic or aquatic habit, caused 

 a restatement of the ' older position by several more recent 

 workers. Thus Dr. Sargent's paper on "Primitive Angio- 

 sperms" {127: 121) still further expands and fortifies the view 

 that dicotyledons retain a more protoangiospermic set of 

 characters than do the monocotyledons. 



The above view also is somewhat strengthened when we 

 compare lists of the earlier angiospermic plants in their two 

 constituents. Thus in Lester Ward's enumeration (U. S. 

 Geol. Surv. 15th Ann. Rep. (1893-94) 313) of plants from 

 beds of cretaceous age, that characterize the Potomac forma- 

 tion of North America, 330 species, or 45 per cent, of the entire 

 recorded flora, are dicotyledonous and few are monocotyled- 

 onous, while the remainder belong to gj^mnosperms or ferns. 

 The above result may be explained in part at least by the 

 usually firmer nature of shrubby or arborescent dicotyledonous 

 than of monocotyledonous leaves. But, granted such, we 

 may next try to ascertain whether any morphological details 

 shown by the primitive ancestors of both can guide us to a 

 correct estimate of their origin. If the monocotyledons did 

 not arise from and in common wdtli the dicotyledons, then 

 we must accept it that we have not a fragment of the ancestors 

 of monocotyledons in a fossil state. 



But, if we accept it that they all sprang from the great cor- 

 daital stock, then we must believe that all had originally two 

 seed-leaves to the embryo, a stem with bundle-ring that in- 

 creased indefinitely, and leaves that were more nearly parallel 

 than net- veined. In other words characters that are now largely 

 separate, and typical of each division, were then conjoined. 

 This also is what we might expect on morphological grounds. 



So, while the Gnetales were evolving and reaching a prob- 

 able climax of generic diversity and abundance during pernio- 

 jurassic times, coeval with them there evidently arose a group 

 of forms that had for a time mainly dicotyledonous tendencies, 



